Liste des publications
Grain size influence on ultrasonic velocities and attenuation
During the last two decades, ultrasonic testing was developed as an efficient tool for materials characterization. Acoustical waves by their passage through materials, carry out a multitude of information contained in the signal on the mechanical and physical properties of the material under inspection. In this paper, an experimental study on steel samples has been performed to study the evolution of some ultrasonic parameters such as wave velocities and attenuation coefficients as function of the steel grains size. The experimental results obtained are discussed and analyzed in order to develop an ultrasonic non-destructive technique to grains size determination. Voir les détails
Mots clés : Ultrasonic, Grain Size, Velocity, Attenuation, Scattering
Time-frequency and wavelet transform applied to selected problems in ultrasonics NDE
In this paper, we contribute by the development of some signal processing in order to enhance the sensibility of flaw detection, to measure thin materials thickness and to characterize defects in nature (planar or volumetric). Features for discrimination of detected echos are extracted in time domain, spectral domain and discrete wavelet representation. Compact feature vector obtained is then classified by different methods: K nearest neighbour algorithm, statistical Bayesian algorithm and artificial neural network. Mallat decomposition algorithm is also developed in order to enhance flaw detectability. Finally, time frequency algorithms based on STFT, Wigner–Ville, Gabor transform are developed and applied to thickness measurements of materials with small thickness. Voir les détails
Mots clés : NDT/NDE, Ultrasonic, time frequency representation, Wavelet transform
Microstructure and mechanical properties of austenitic stainless steels after cold rolling
The effect of austenite stability on the evolution of microstructure and mechanical properties of three austenitic stainless steels during cold rolling has been studied. Samples of different grain sizes have been used to characterize the microstructures during deformation. In the case of 304/8% Ni and 304/10% Ni stainless steels, the transformation microstructures consist of mechanical twins: ε-martensite and α′-martensite. No hexagonal close-packed (hcp) ε-martensite was detected in 316 stainless steel. The volume fraction of α′-martensite formed increases with increasing strain in 304 and 316 stainless steels for a given grain size. The amount of α′ phase increases with a decrease in grain size in 304 stainless steel, while the formation of this phase has been found to be grain size insensitive in 316 stainless steel. The strain-hardening behavior exhibited by the three stainless steels used in this study indicates the contribution of both α′-martensite and grain size strengthening in the case of both 304 stainless steels, while only grain size contribution was found in the case of 316 stainless steel. Voir les détails
Mots clés : deformation twinning, slip, stacking fault energy, strain-induced martensite
Elaboration of some signal processing algorithms in ultrasonic techniques: application to materials NDT
In ultrasonic techniques, information on defect characterization possibilities has required more evolved technique development than classical methods. To obtain a high probability of defect detection, these methods use signal-processing algorithms in order to enhance the signal-to-noise ratio. These methods are also used to discriminate between planar and volumetric defects. In this paper, some signal-processing algorithms are developed and implemented on a computer to allow their utilization in real-time processing of ultrasonics NDT results. Voir les détails
Mots clés : Cross-correlation function, Hilbert transform, NDT, Split spectrum techniques, Ultrasonics
Thermally affected characterization region by Barkhausen noise
The controlling of some industrial components require the development of new and particular nondestructive testing techniques. The testing method using Barkhausen noise (BN) is a particular one which can be applied to ferromagnetic materials. It is a magnetic nondestructive evaluation method and can provide very important information about the material structure. The aim of our work is to study the material structure using this technique to characterize the region submitted to thermal processing. Samples of steel have been heated at temperatures between 650°C and 1200°C with variable parameters (time processing, maintenance time, etc.). Acoustic BN processing allows an easy interpretation of results. Micrographs of samples have been obtained to confirm the results obtained by BN. Voir les détails
Mots clés : Non-Destructive testing, Barkhausen noise, Fast Fourier transformer, Magnetic field, welding, microstructure
Ultrasonic characterization of materials hardness
In this paper, an experimental technique has been developed to measure velocities and attenuation of ultrasonic waves through a steel with a variable hardness. A correlation between ultrasonic measurements and steel hardness was investigated. Voir les détails
Mots clés : Ultrasounds, hardness, Velocity, Attenuation
CESOP: Computer Aided Ultrasonic Testing with Expertise and Arc Welding Parameters Optimization
In metallic constructions, the assemblage techniques are in continual evolution dealing with the technological progress. However one can not insure the quality of an assemblage without proceeding to a quality control. A comparative study on DDT methods, presented during the SMIRT10 in Montrey 1989 has shown that the radiography appears better than the manual ultrasonic testing for the detection of almost welding defects type. Neither it is efficient for the diagnostic, the manual ultrasonic testing gives performances about 50% of detection. On the other hand, the mechanized or automatic ultrasonic testing, insures a probability of detection for plan or volume defects in the order of 80%.On the basis of these estimations, we have judged that it would be judicious to encourage the automatic ultrasonic testing. For this purpose and in the goal to approach the ultrasonic testing method by the radiography on adopting the imagery and the signal processing to the ultrasonic techniques, we present the system, that allows the interpretation automation of the ultrasonic signals identified (1), and simulates the recognition of the defect shape according to the radiography principle (2). Results of this interpretation, provided by << ARCUS >> will be confided to the expert system << SYSO >> (3) for the diagnostic of the defects causes. The exploitation of<< SYSO >> has shown that, a bad determination of welding parameters is the most current defect cause. The << OPTIMA >> system, modeled on the computer methods basis, help to the optimization of the arc welding parameters. These three [3] processes << ARCUS >>, << SYSO>> and << OPTIMA >> cooperate through coupling interfaces to create the: Computer aided Ultrasonic Testing, with Expertise and Arc Welding Parameters Optimization, called << CESOP >>,whose purpose is improving the arc welding quality and increasing as well as the security of the assembled construction. Voir les détails
Mots clés : Welds, During Manufacture
Abstract factor analysis of spectroscopic data-another statistical fallacy
It is shown that statistical arguments which have been used in published work to correct absorption spectra for medium effects are mathematically unsound and chemically misleading. Understanding this misconception, however, is helpful tomake clear the limits of abstract factor analysis in chemistry. Voir les détails
Mots clés : absorption spectra, abstract factor analysis